We propose to test two linked hypotheses regarding children with short bowel syndrome. First, we propose to test the widely held assumption that children with short bowel syndrome have a high incidence of bacterial overgrowth in the remaining small bowel. Secondly, we propose to test the hypotheses that changing the resident flora in the intestine of such patients will result in a decrease in fat loss in the stool, a change in stool flora and an improvement in the evidence of hepatic toxicity that affects most of these patients.